[nailgun] Root size calculation blocks deployment on non-first drive in CentOS

Bug #1266569 reported by Dmitry Borodaenko
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Fuel for OpenStack
Won't Fix
Medium
Fuel Python (Deprecated)

Bug Description

When deploying CentOS on a second drive that is significantly smaller than the first drive, deployment is blocked because minimum root partition size gets set to a size larger than the second drive. There's more than one problem with the way root size is calculated in _calc_root_size():

1) Hardcoded assumption that the size of the _first_ drive in the system is relevant to determining the OS VG size (not true on CentOS where the OS volume group can go on any drive or even span multiple drives).

2) Hardcoded assumption that drives smaller than 50GB are not going to be used to install OS (not true for customers who want to put OS on a small SSD).

3) Confusion between the _default_ root size and _minimum_ root size. Minimum root size should still be 10GB (or whatever amount of disk space is required to install _all_ Mirantis OpenStack components with all options enabled). Default size may be larger than that but should not be used to validate disk sizes before deployment.

This problem was brought up in a code review but didn't get addressed before the change was merged:
https://review.openstack.org/#/c/56322/6/nailgun/nailgun/volumes/manager.py

Tags: nailgun
Andrew Woodward (xarses)
Changed in fuel:
status: New → Triaged
Evgeniy L (rustyrobot)
Changed in fuel:
assignee: Fuel Library Team (fuel-library) → Fuel Python Team (fuel-python)
Mike Scherbakov (mihgen)
Changed in fuel:
milestone: 4.1 → 5.0
Dmitry Pyzhov (dpyzhov)
Changed in fuel:
milestone: 5.0 → 5.1
Dmitry Ilyin (idv1985)
summary: - Root size calculation blocks deployment on non-first drive in CentOS
+ [nailgun] Root size calculation blocks deployment on non-first drive in
+ CentOS
Dmitry Pyzhov (dpyzhov)
Changed in fuel:
milestone: 5.1 → 6.0
status: Triaged → Won't Fix
Revision history for this message
Andrew Woodward (xarses) wrote :

Please explain Won't Fix status

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Related blueprints

Remote bug watches

Bug watches keep track of this bug in other bug trackers.